Testing the accuracy of low-cost data streams for determining single-person office occupancy and their use for energy reduction of building services

Abstract

We explored methods of detecting occupancy in single-person offices using data already collected by the occupant’s PC, or data from relatively cheap sensors added to the PC. We collected data at 15-s intervals for up to 31 days in each of 28 offices. A combination of low/no cost sensors (webcam-based motion detection, and keyboard and mouse activity) was much more accurate at detecting occupancy than a commercial ceiling-based passive infrared (PIR) sensor, and provided overall daytime accuracy >90%, with very low false negative rates. This enhanced detection performance would enable a reduction in the timeout periods for building service curtailment on space vacancy. For example, lighting switch-off timeout could be reduced from the current energy code standard of 20min to less than 5min, increasing energy savings potential by 25–45%. We then deployed this system in a proof-of-concept demonstration, using it to control lighting, heating, ventilation, and air conditioning (HVAC), and plug loads in a mock-up office environment. Tests were run over nine occupied days (six in cooling season, three in heating season). The system delivered energy savings of 15–68%, with no reported false negative errors.